The Financial Access Initiative’s goal is to improve understandings of the ways that low-income households demand and use financial services—and to inform innovation and policy reform. We generate reliable, independent research, and translate findings into practical tools for those making decisions that affect financial access for the poor.

FAI research provides a framework for decision-making built around the economics and psychology of low-income households. The framework informs the identification of puzzles, tensions, and opportunities in developing inclusive financial systems. The initiative aims to inform policy development necessary for continuing the rapid expansion of financial access for the under-served poor. This expansion will require process and product innovations, regulatory reforms, and expanded financial intermediation.

FAI empirical research begins by systematizing what we know and don’t know around key issues of demand and impact. Then, based on the knowledge gaps identified, a group of linked research projects generates evidence on these critical but unresolved issues. Third, and perhaps most important, a communications effort ensures that the knowledge generated through our research helps directly inform decisions.

Research initiatives are led by Jonathan Morduch, Dean Karlan and Sendhil Mullainathan. FAI's field research is conducted by Innovations for Poverty Action.

Research Areas

Households & Demand

The most basic policy objective is to expand financial access. In order to do this, it is necessary to gain a more comprehensive understanding of the dimension and nature of needs and the services or products that best fulfill them.

How much do prices matter to customers – or is gaining financial access far more critical? How does psychology affect decision-making? Why are take-up rates often so low for new products? Does education and financial literacy make a difference?

Impact

One of the most difficult effects to quantify is the impact that microfinance services are having on the poor. Shops could be expanded, livestock purchased, livelihoods stabilized.  Success stories are often told, but are not always typical. FAI research establishes the extent of average impacts of financial services using rigorous, statistical methods to measure causal relationships. Focuses are on the roles of new credit, savings, and insurance products.

Markets & Institutions

The microfinance institutional landscape is constantly changing with the range of institutions ranging from private commercial banks to non-profit organizations.  Growing interest and participation by commercial investors and financial institutions could impact the type of services and the scale of access. FAI examines cross-country evidence to present a empirical evidence on the nature and impact of the changing conditions on institutions and households alike.

Regulation & Policy

Regulators of financial services do not always have the benefit of learning from their peers in other countries.  The objective of FAI research is to document examples of regulators safeguarding the stability of the financial sector while at the same time enabling the flexibility needed by institutions to expand services to the un-banked.  The final result will be a practical tool based on past experiences, hurdles, and possibilities that can be used to forge appropriate regulatory regimes.

FAI continues to develop new collaborative relationships to broaden the potential of the research and dissemination of findings. Click here for more information on becoming an FAI research partner.

Randomized Control Trials

FAI’s field research focuses on the use of randomized control trials (RCTs), methods pioneered in health and medical research. In RCTs, research participants are randomly assigned to a group that receives the intervention or a group that does not receive the intervention. This method is the most effective way of assuring that comparisons are not biased by effects other than the intervention in question (quantitatively important biases can stem from macroeconomic and regional variables, environmental shocks, and various selection biases). The difference in outcomes between treatment and control groups yields a clean measure of the effect of the intervention.